North Wales is about to sit at the centre of one of the most important economic experiments in modern Welsh policy.
The proposed AI Growth Zone spanning Prosperity Parc on Anglesey and Trawsfynydd in Gwynedd is part of a much bigger push to make Wales one of the UK’s main testbeds for AI-driven growth. It will attract attention because of infrastructure, energy and investment. But the real question is not whether North Wales can host major AI infrastructure. It is whether that infrastructure will create practical value that reaches the wider North Wales economy.
That distinction matters. The big numbers will always be tempting. Across the two Welsh Growth Zones, the public story is likely to focus on investment, power demand, construction activity and headline jobs. In North Wales, projections point to more than 3,400 jobs. But as the wider framework argues, many of the largest early employment effects will come during construction and build-out. The longer-term prize is different: more productive SMEs, more specialist suppliers, more founder-led firms and more routes into high-value work rooted in the local economy.
That is especially important in Wales because the underlying economic challenge is not a shortage of headlines. It is a shortage of productivity growth. Wales has the lowest productivity of any UK nation or region, while SMEs account for around 62.3% of Welsh employment. If AI policy does not improve outcomes for smaller firms, it will not move the Welsh economy in a serious way. The implication is simple and uncomfortable: if AI does not reach tens of thousands of small firms, it will not fix the productivity problem, no matter how many data centres get built.
So what should success mean in North Wales?
First, it should mean that infrastructure deals come with visible local obligations. If North Wales is offering accelerated planning, better grid access, discounted electricity or any other public support to major operators, then Welsh firms should see something concrete in return. That means published targets for spend with Welsh suppliers, a clear share of work for Welsh SMEs, transparent reporting on procurement and a serious effort to develop local specialist capability rather than importing everything from outside the region. A Growth Zone cannot count as a local success if the steel, software, security, professional services, maintenance and follow-on contracts mostly bypass North Wales businesses.
Second, success should mean real AI adoption in North Wales firms, not just a few showcase pilots. The framework is clear that each Growth Zone will come with adoption and skills funding, and that the risk is superficial activity: short-run trials, generic bootcamps and events that look busy but do not change how firms operate. In North Wales, success would look more like 6 to 12 month projects with real businesses in sectors that matter locally: manufacturing, construction, tourism, logistics, food and drink, and related specialist services. It would mean firms redesigning workflows, adopting tools that improve output and quality, and creating new AI-enabled offers of their own. It would also mean measuring results properly rather than counting attendance and calling it impact.
Third, North Wales should be where the rural and Welsh-language dimension is taken seriously rather than treated as a footnote. One of the clearest warnings in the paper is that rural SMEs face additional barriers around connectivity, energy, skills and language. That matters in North Wales more than almost anywhere else. If this Growth Zone ends up helping only a narrow set of urban or already digitally confident firms, it will miss one of the most important tests of whether AI-led growth can work for Wales as it actually exists. Success in North Wales should therefore include strong rural participation, support for Welsh-language AI tools and interfaces, and delivery models that do not assume every business can navigate a fragmented innovation landscape on its own.
Fourth, success should mean that energy access improves for smaller firms rather than tightening around them. The framework’s third test is really about fairness as much as infrastructure. If AI Growth Zones get priority treatment for grid access while local firms are left waiting years to electrify vehicles, heat premises or expand production, then the result will be an energy island: impressive on paper, frustrating in practice. In North Wales, success should mean ring-fenced local capacity for SMEs and community infrastructure, more flexible connection models for smaller users, and a clear expectation that major projects help anchor renewables and storage that benefit nearby businesses as well as large sites.
Fifth, success should be judged by diffusion, not just presence. The wrong question is whether North Wales can attract AI infrastructure. It can. The better question is whether the benefits spread outward into the local business base, local supply chains, public institutions and labour market. Are local SMEs finding it easier to win work, adopt tools, hire skilled people and start specialist firms? Are colleges, universities, councils and health bodies plugged into the opportunity in a practical way? Are more people in North Wales able to move into higher-value technical and commercial roles because the Growth Zone is there? Those are the questions that tell you whether a Growth Zone belongs to a place rather than simply occupies it.
A different scorecard for North Wales
That, in turn, suggests a different scorecard for North Wales. Instead of stopping at jobs, capital spend and press releases, success metrics should include the share of procurement going to Welsh SMEs, the number of North Wales firms completing multi-month AI projects, productivity gains achieved in those firms, the number of new specialist suppliers created or scaled, and whether energy and skills access have improved for smaller businesses. The wider framework also argues for a public “Growth Zone SME Compact” so firms know what support, routes to supply and entry points into adoption programmes they can expect. North Wales should be the place where that compact becomes real.
There is a bigger opportunity here. Done badly, the North Wales AI Growth Zone could become another example of impressive infrastructure with thin local spillover. Done well, it could show what an SME-first model actually looks like: targeted procurement, practical adoption support, bilingual and rural inclusion, better energy access, and a stronger local ecosystem of firms that can grow because the infrastructure is there. That is the standard worth using.
North Wales does not need to prove that it can host the future.
It needs to prove that the future can be made useful to the businesses already there.